Metraipolis › archive › dc_c713ffbc3290

SepRQ : Self-Supervised Speech Mixture Representation Learning via Mask-Free, Multi-Scale Source Separation — arXiv:2610.04690

Source
arXiv
Original
https://arxiv.org/abs/2610.04690
License (as archived)
CC0-1.0
Archived
2026-10-06T05:28:47.210Z
SHA-256 of the archived text
84c6ea4c7603682c10d3bce838dcd1437cff8d6e8116359ff567aaa164434d90
Stable URL of this page
https://metraipolis.com/documento/dc_c713ffbc3290

How to cite

SepRQ : Self-Supervised Speech Mixture Representation Learning via Mask-Free, Multi-Scale Source Separation — arXiv:2610.04690 — arXiv, https://arxiv.org/abs/2610.04690 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c713ffbc3290 (Metraipolis archive dc_c713ffbc3290, sha256 84c6ea4c7603682c10d3bce838dcd1437cff8d6e8116359ff567aaa164434d90)
BibTeX
@misc{dcc713ffbc3290,
  title = {SepRQ : Self-Supervised Speech Mixture Representation Learning via Mask-Free, Multi-Scale Source Separation — arXiv:2610.04690},
  howpublished = {arXiv},
  url = {https://arxiv.org/abs/2610.04690},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c713ffbc3290; sha256 84c6ea4c7603682c10d3bce838dcd1437cff8d6e8116359ff567aaa164434d90},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.04690. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Self-supervised learning (SSL) is standard for speech representation learning, but mainstream models are designed around single-speaker audio, limiting their usefulness in multi-speakers scenarios. We present SepRQ, an open-source SSL framework that replaces masked prediction with a pseudo-source-separation objective over frozen random-projection codebooks. By adopting a novel mask-free, multiresolution approach, SepRQ achieves state-of-the-art performance in Speaker Diarization and Speech Separation on the SUPERB benchmark, surpassing WavLM and other cocktail-party derived SSLs at both Base and Large scales, while requiring only 85.68M inference parameters. SepRQ also demonstrates strong performance across target-speaker tasks requiring enrollment (such as Target-Speaker Automatic Speech Recognition), and on the challenging multi-domain DIHARD 3 diarization dataset. Notably, we report strong separation capabilities on three-speaker mixtures (WSJ0-3Mix), where current SSL literature struggles. While cocktail-party SSLs remain scarce and closed-source, limited to C-HuBERT and the enrollment-based SA-WavLM, we open-source SepRQ to the community.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno).
Categorie arXiv: Audio and Speech Processing (eess.AS) ; Computation and Language (cs.CL); Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2610.04690 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-06T05:24:20.423Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).